Agency Th eory 37
Furthermore, Maynard-Moody and Musheno (2009) reveal that the set of
relationships among citizens, agents, and the state is more complex than previ-
ously thought, which makes controlling bureaucracy more diffi cult. Th ese fi nd-
ings are consistent with others, who agree that street-level discretion reduces
the chances of control over those street-level workers (Brehm and Gates 1997).
Maynard-Moody and Musheno fi nd, for example, that, although we should ac-
knowledge the diffi culties of street-level work and tolerate some discretion in de-
cisionmaking, we must not forget that street-level bureaucrats are indeed agents
of the state. Th is creates, they argue, an unresolvable tension between the ex-
pectations of principal-agent theory and the relationships between citizens and
street-level agents: should street-level bureaucrats respond directly to citizen
preferences, or should they respond to the preferences of elected offi cials, who
represent the citizenry?
Nevertheless, in their review of fi ndings based on agency theory, Wood and
Waterman (1994) indicate the following:
- Bureaucratic responsiveness to political control is the norm rather than
the exception. A wide range of contingent factors, such as time, pres-
idential and congressional agreement, and many others, infl uence the
degree of bureaucratic responsiveness. - Political control mechanisms are important, especially presidential ap-
pointments, congressional appropriations power, hearings, and con-
gressional staff eff ectiveness. - Organization matters. Agencies in executive or cabinet departments
are more responsive, whereas independent agencies are less so. - Presidential statements are infl uential, as are the statements of senior
congressional leaders.
How do bureaucracies respond to political control? In their review of agency
theory research of federal agencies, Wood and Waterman conclude that there are
dynamic bidirectional relationships in which legislators signal preferences to bu-
reaucrats and bureaucrats signal preferences to legislators:
Some would argue that such two-way power relations are evidence of political
dysfunction, given that bureaucracies are nonelected institutions. However, we
suggest that the opposite is true. It is healthy for bureaucracy to use its infor-
mation advantages to better inform principals on either policy matters or the
nature of the bureaucratic process. Furthermore, bureaucratic resistance to duly
elected politicians may actually sometimes be more consistent with democracy
and public preferences than bureaucratic responsiveness may be. Citizens and
many groups strongly opposed the de-regulation of the environment advocated
by the Reagan administration, and the bureaucracy served as a check on presi-
dential power. (1994, 126)